AI News Archive: May 18, 2026 — Part 15
Sourced from 500+ daily AI sources, scored by relevance.
- All the Juiciest Evidence From the Blockbuster OpenAI Trial
We sifted through evidence and testimony to figure out how billionaires Elon Musk, Sam Altman and Greg Brockman ended up in a courtroom.
- Musk’s OpenAI Claims Rejected by Jury
Plus, Trump says he’ll pause plans to attack Iran, and tech founders attend etiquette classes
- Key moments in the Musk vs OpenAI trial
Key moments in the Musk vs OpenAI trial Reuters
- Instant View: California jury sides with OpenAI over Musk lawsuit
Instant View: California jury sides with OpenAI over Musk lawsuit Reuters
- Elon Musk Loses Case Against Sam Altman Over OpenAI’s Overhaul
A jury rejected Elon Musk’s claims that OpenAI under Sam Altman’s leadership betrayed its mission to benefit the public by morphing into a for-profit business, finding that he waited too long to sue the company.
- Elon Musk loses legal fight against OpenAI
Elon Musk loses legal fight against OpenAI The Telegraph
- Elon Musk Loses Landmark Lawsuit Against OpenAI
The nine-member panel took only two hours to return a verdict in favor of OpenAI on Monday, which the judge quickly adopted as her own final decision.
- Elon Musk took too long to sue OpenAI, jury unanimously agrees
Musk plans to appeal after judge immediately affirmed the jury's decision.
- Elon Musk has lost his lawsuit against Sam Altman and OpenAI
Elon Musk's claim that he was mistreated by his OpenAI co-founders failed after nine California jurors decided in a unanimous verdict that his lawsuits had been filed too late.
- Shallow ReLU$^s$ Networks in $L^p$-Type and Sobolev Spaces: Approximation and Path-Norm Controlled Generalization
We study approximation by shallow ReLU$^s$ networks, $σ_s(t)=\max{0,t}^s$, and the generalization behavior of such networks under $\ell_1$ path-norm control. For the $L^p$-type integral spaces $\widetilde{\mathcal{F}}_{p,τ_d,s}$, $1\le p\le2$, we establish approximation bounds for shallow networks u...
- Improved Baselines with Representation Autoencoders
Representation Autoencoders (RAE) replace traditional VAE with pretrained vision encoders. In this paper, we systematically investigate several design choices and find three insights which simplify and improve RAE. First, we study a generalized formulation where the representation is defined as sum ...
- Anthropic has acquired the dev tools startup used by OpenAI, Google, and Cloudflare
Stainless, a New York-based startup, founded in 2022, rose to prominence in the emerging AI industry for automating the creation and maintenance of software development kits, or SDKs — the libraries developers use to interact with APIs.
- Musk loses OpenAI court battle after jury finds he waited too long to sue
Jurors spent weeks hearing about Musk's claim that Altman had "stolen a charity."
- Code as Agent Harness
Recent large language models (LLMs) have demonstrated strong capabilities in understanding and generating code, from competitive programming to repository-level software engineering. In emerging agentic systems, code is no longer only a target output. It increasingly serves as an operational substra...
- Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation
Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding, where answers often depend on small but decisive evidence in the full image. We observe a regional-to-global perception gap: the same MLLM answers fine-grained questions more accurately when conditioned o...
- Predictable Confabulations: Factual Recall by LLMs Scales with Model Size and Topic Frequency
While scaling laws govern aggregate large language model performance, no scaling law has linked factual recall to both model size and training-data composition. We evaluated 38 models on over 8,900 scholarly references evaluated by an automated reference verification system. Recall quality follows a...
- Semantic Generative Tuning for Unified Multimodal Models
Unified multimodal models (UMMs) strive to consolidate visual understanding and visual generation within a single architecture. However, prevailing training paradigms independently optimize understanding via sparse text signals and generation through dense pixel objectives. Such a decoupled strategy...
- Federated Martingale Posterior Samping
Federated Bayesian neural networks require fixing a prior on the model parameters together with a likelihood. Eliciting meaningful priors on the weight space of modern overparameterized models is notoriously difficult, and misspecification of either component can severely degrade accuracy and calibr...
- Distilling Tabular Foundation Models for Structured Health Data
Tabular foundation models (TFMs) achieve strong performance on health datasets, but their inference cost and infrastructure requirements limit practical use. We study whether their predictive behavior can be transferred to lightweight tabular models through knowledge distillation. Since in-context T...
- SkillGenBench: Benchmarking Skill Generation Pipelines for LLM Agents
As LLM agents are increasingly built around reusable skills, a central challenge is no longer only whether agents can use provided skills, but whether they can generate correct, reusable, and executable skills from repositories and documents. Existing benchmarks primarily evaluate the efficacy of gi...
- Democratizing Large-Scale Re-Optimization with LLM-Guided Model Patches
Optimization models developed by operations research (OR) experts are often deployed as decision-support systems in industrial settings. However, real-world environments are dynamic, with evolving business rules, previously overlooked constraints, and unforeseen perturbations. In such contexts, end ...
- Learning Quantifiable Visual Explanations Without Ground-Truth
Explainable AI (XAI) techniques are increasingly important for the validation and responsible use of modern deep learning models, but are difficult to evaluate due to the lack of good ground-truth to compare against. We propose a framework that serves as a quantifiable metric for the quality of XAI ...
- Lance: Unified Multimodal Modeling by Multi-Task Synergy
We present Lance, a lightweight native unified model supporting multimodal understanding, generation, and editing for both images and videos. Rather than relying on model capacity scaling or text-image-dominant designs, Lance explores a practical paradigm for unified multimodal modeling via collabor...
- COOPO: Cyclic Offline-Online Policy Optimization Algorithm
Offline reinforcement learning struggles with distributional shift and constrained performance due to static dataset limitations, while online RL demands prohibitive environment interactions. The recent advent of hybrid offline-to-online methods bridges these domains but suffers from distribution dr...
- Efficient Lookahead Encoding and Abstracted Width for Learning General Policies in Classical Planning
Generalized planning aims to learn policies that generalize across collections of instances within a classical planning domain. Recent Graph Neural Network (GNN) approaches have learned nearly perfect policies for several domains. This work improves on the recently published idea of Iterated Width (...
- Position: A Three-Layer Probabilistic Assume-Guarantee Architecture Is Structurally Required for Safe LLM Agent Deployment
This position paper argues that enforcing LLM agent safety within a single abstraction layer is not merely suboptimal but categorically insufficient for deployed LLM agents -- a structural consequence of how agent execution works, not a contingent limitation of current systems. The three dimensions ...
- GIM: Evaluating models via tasks that integrate multiple cognitive domains
As LLM benchmarks saturate, the evaluation community has pursued two strategies to increase difficulty: escalating knowledge demands (GPQA, HLE) or removing knowledge entirely in favor of abstract reasoning (ARC-AGI). The first conflates memorization with capability; the second divorces reasoning fr...
- KairosHope: A Next-Generation Time-Series Foundation Model for Specialized Classification via Dual-Memory Architecture
Time Series Foundation Models (TSFMs) have demonstrated notable success in general-purpose forecasting tasks; however, their adaptation to specialized classification problems remains constrained by the computational bottleneck of standard attention and the systematic omission of classical statistica...
- Statistical Limits and Efficient Algorithms for Differentially Private Federated Learning
Federated Learning is a leading framework for training ML and AI models collaboratively across numerous user devices or databases. We study the trade-offs among estimation accuracy, privacy constraints, and communication cost for differentially private (DP) federated M estimation. The two standard m...
- Pocket Foundation Models: Distilling TFMs into CPU-Ready Gradient-Boosted Trees
A fraud scorer needs to answer in under 2 ms. The best tabular foundation models (TFMs) take 151-1,275 ms on GPU. We close this gap by distilling the TFM offline into an XGBoost or CatBoost student that runs natively on CPU. The central obstacle is specific to in-context learning (ICL) teachers: the...
- Post-Trained MoE Can Skip Half Experts via Self-Distillation
Mixture-of-Experts (MoE) scales language models efficiently through sparse expert activation, and its dynamic variant further reduces computation by adjusting the activated experts in an input-dependent manner. Existing dynamic MoE methods usually rely on pre-training from scratch or task-specific a...
- Data Presentation Over Architecture: Resampling Strategies for Credit Risk Prediction with Tabular Foundation Models
Credit default prediction is a tabular learning problem with severe class imbalance, heterogeneous features, and tight latency budgets. Tabular Foundation Models (TFMs) approach this problem through in-context learning, which makes their predictions sensitive to how the context window is built. We b...
- Position: Weight Space Should Be a First-Class Generative AI Modality
Neural network checkpoints have quietly become a large-scale data resource: millions of trained weight vectors now exist, each encoding task-, domain-, and architecture-specific knowledge. This position paper argues that model checkpoints should be treated as a first-class data modality, and that ge...
- Attention-based PCA
We study attention mechanisms through the lens of a canonical unsupervised problem: principal component analysis (PCA). We show that, when trained on Gaussian data, both softmax and linear attention layers learn parameters that align with the principal eigenvectors of the covariance matrix, thereby ...
- SCICONVBENCH: Benchmarking LLMs on Multi-Turn Clarification for Task Formulation in Computational Science
Large Language Models (LLMs) are increasingly deployed as scientific AI as- sistants, and a growing body of benchmarks evaluates their capabilities across knowledge retrieval, reasoning, code generation, and tool use. These evaluations, however, typically assume the scientific problem is already wel...
- CrossView Suite: Harnessing Cross-view Spatial Intelligence of MLLMs with Dataset, Model and Benchmark
Spatial intelligence requires multimodal large language models (MLLMs) to move beyond single-view perception and reason consistently about objects, visibility, geometry, and interactions across multiple viewpoints. However, progress in cross-view reasoning remains limited by three major gaps: the sc...
- Stochastic Penalty-Barrier Methods for Constrained Machine Learning
Constrained machine learning enables fairness-aware training, physics-informed neural networks, and integration of symbolic domain knowledge into statistical models. Despite its practical importance, no general method exists for the non-convex, non-smooth, stochastic setting that arises naturally in...
- CATA: Continual Machine Unlearning via Conflict-Averse Task Arithmetic
Vision-language models (VLMs) have shown remarkable ability in aligning visual and textual representations, enabling a wide range of multimodal applications. However, their large-scale training data inevitably raises concerns about privacy, copyright, and undesirable content, creating a strong need ...
- Latent Action Reparameterization for Efficient Agent Inference
Large language model (LLM) agents often rely on long sequences of low-level textual actions, resulting in large effective decision horizons and high inference cost. While prior work has focused on improving inference efficiency through system-level optimizations or prompt engineering, we argue that ...
- Not What You Asked For: Typographic Attacks in Household Robot Manipulation
Open-vocabulary embodied AI agents increasingly rely on vision-language models such as CLIP for object perception and task grounding. However, the shared embedding space that enables this flexibility introduces a structural vulnerability to typographic attacks, where printed text in a physical scene...
- AMARIS: A Memory-Augmented Rubric Improvement System for Rubric-Based Reinforcement Learning
Rubric-based reward shaping is an effective method for fine-tuning LLMs via RL, where structured rubrics decompose standard outcome rewards into multiple dimensions to provide richer reward signals. Recent works make the rubrics adaptive based on local signals such as the rollouts from the current s...
- Estimating Item Difficulty with Large Language Models as Experts
Accurate estimates of item difficulty are essential for valid assessment and effective adaptive learning. However, for newly created tasks, response data are typically unavailable. Pretesting and expert judgement can be costly and slow, while machine learning methods often require large labelled tra...
- STT-Arena: A More Realistic Environment for Tool-Using with Spatio-Temporal Dynamics
Large language models (LLMs) deployed in real-world agentic applications must be capable of replanning and adapting when mid-task disruptions invalidate their prior decisions. Existing dynamic benchmarks primarily measure whether LLMs can detect temporal changes in a timely manner, leaving the compl...
- DashAttention: Differentiable and Adaptive Sparse Hierarchical Attention
Current hierarchical attention methods, such as NSA and InfLLMv2, select the top-k relevant key-value (KV) blocks based on coarse attention scores and subsequently apply fine-grained softmax attention on the selected tokens. However, the top-k operation assumes the number of relevant tokens for any ...
- ESI-Bench: Towards Embodied Spatial Intelligence that Closes the Perception-Action Loop
Spatial intelligence unfolds through a perception-action loop: agents act to acquire observations, and reason about how observations vary as a function of action. Rather than passively processing what is seen, they actively uncover what is unseen - occluded structure, dynamics, containment, and func...
- Actionable World Representation
Inspired by the emergent behaviors in large language models that generalized human intelligence, the research community is pursuing similar emergent capabilities within world models, with a emphasis on modeling the physical world. Within the scope of physical world model, objects are the fundamental...
- What Does the AI Doctor Value? Auditing Pluralism in the Clinical Ethics of Language Models
Medicine is inherently pluralistic. Principles such as autonomy, beneficence, nonmaleficence, and justice routinely conflict, and such ethical dilemmas often sharply divide reasonable physicians. Good clinical practice navigates these tensions in concert with each patient's values rather than imposi...
- DexHoldem: Playing Texas Hold'em with Dexterous Embodied System
Evaluating embodied systems on real dexterous hardware requires more than isolated primitive skills: an agent must perceive a changing tabletop scene, choose a context-appropriate action, execute it with a dexterous hand, and leave the scene usable for later decisions. We introduce DexHoldem, a real...
- PopPy: Opportunistically Exploiting Parallelism in Python Compound AI Applications
Compound AI applications, which compose calls to ML models using a general-purpose programming language like Python, are widely used for a variety of user-facing tasks, from software engineering to enterprise automation, making their end-to-end latency a critical bottleneck. In contrast to tradition...
- Ensembling Tabular Foundation Models - A Diversity Ceiling And A Calibration Trap
Tabular foundation models (TFMs) now match or beat tuned gradient-boosted trees on a growing fraction of tabular tasks, but no single TFM wins on every dataset. Ensembling is the go to fix here, and it works less well than expected. Six modern TFMs form a near-redundant pool: their mean pairwise Q-s...